RT Journal Article
SR Electronic
T1 Clinical data specification and coding for cross-analyses with omics data in autoimmune disease trials
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 360719
DO 10.1101/360719
A1 Lorenzon Roberta
A1 Drakos Iannis
A1 Claire Ribet
A1 Sophie Harris
A1 Cordoba Maeva
A1 Tran Olivia
A1 Dasque Eric
A1 Cacoub Patrice
A1 Hartemann Agnes
A1 Bodaghi Bahram
A1 Saadoun David
A1 Berenbaum Francis
A1 Grateau Gilles
A1 Ronco Pierre
A1 Benveniste Olivier
A1 Mariampillai Kuberaka
A1 Sellam Jeremie
A1 Seksik Philippe
A1 Rosenzwajg Michelle
A1 Six Adrien
A1 Bernard Claude
A1 Aheng Caroline
A1 Vicaut Eric
A1 Klatzmann David
A1 Mariotti-Ferrandiz Encarnita
YR 2018
UL http://biorxiv.org/content/early/2018/07/05/360719.abstract
AB Objectives Autoimmune and inflammatory diseases (AIDs) form a continuum of autoimmune and inflammatory diseases, yet AIDs’ nosology is based on syndromic classification. The TRANSIMMUNOM trial (NCT02466217) was designed to re-evaluate AIDs nosology through clinic-biological and multi-omics investigations of patients with one of 19 selected AIDs. To allow cross-analyses of clinic-biological data together with omics data, we needed to integrate clinical data in a harmonized database.Materials and Methods We assembled a clinical expert consortium (CEC) to select relevant clinic-biological features to be collected for all patients and a cohort management team comprising biologists, clinicians and computer scientists to design an electronic case report form (eCRF). The eCRF design and implementation has been done on OpenClinica, an open-source CFR-part 11 compliant electronic data capture system.Results The CEC selected 865 clinical and biological parameters. The CMT selected coded the items using CDISC standards into 5835 coded values organized in 28 structured eCRFs. Examples of such coding are check boxes for clinical investigation, numerical values with units, disease scores as a result of an automated calculations, and coding of possible treatment formulas, doses and dosage regimens per disease.Discussion 21 CRFs were designed using OpenClinica v3.14 capturing the 5835 coded values per patients. Technical adjustment have been implemented to allow data entry and extraction of this amount of data, rarely achieved in classical eCRFs designs.Conclusions A multidisciplinary endeavour offers complete and harmonized CRFs for AID clinical investigations that are used in TRANSIMMUNOM and will benefit translational research team.